Organizers of a leading neuroscience conference are trying to allay concerns after a policy that bars submissions from researchers based at institutions sanctioned by the United States government attracted intense scrutiny online.
Representatives of the 40th Annual Conference on Neural Information Processing Systems (NeurIPS), due to take place in Sydney, Australia, in December 2026, took to X (formerly Twitter) on 27 March to apologize and note that there had been miscommunication between the NeurIPS Foundation, which runs the conference, and its legal team, which led to confusion and anger among researchers.
Specifically, the confusion was around the extent to which the policy would restrict researchers based at U.S.-sanctioned institutions in publishing, editing or reviewing papers at NeurIPS. The conference’s main track handbook, released on 23 March, also contained a link to the sanctions search tool created by the U.S. Treasury Department’s Office of Foreign Assets Control (OFAC) that covers a broader set of restrictions than those NeurIPS is required to follow.
But text on the NeurIPS website has since been tweaked, and the link has since been updated to point to an OFAC webpage listing people and companies that are sanctioned by the U.S.
“We have updated the link and clarified the text of our policy, which is consistent with that of … other international conferences and NeurIPS in the past,” the conference’s organizers wrote on X and in a statement shared with The Transmitter. “As in previous years, NeurIPS welcomes submissions from all compliant institutions and individuals.”
The previous wording had caused anger among researchers, who voiced concerns on social media, arguing that the requirement was unfair and suppressed academic freedom. As a result, some had called for a boycott of NeurIPS, and others had suggested the NeurIPS Foundation should move outside U.S. jurisdiction.
If implemented with the previous wording, the policy would have been “pointless and self-destructive,” says Anthony Zador, Alle Davis and Maxine Harrison Professor of Neurosciences at Cold Spring Harbor Laboratory.
T
Yasser Roudi, professor of disordered systems at King’s College London, says the NeurIPS handbook could have included information about whether the conference organizers have contacted OFAC asking for exemptions or waivers for sanctioned institutions. “It’s just useful to communicate that they have tried and failed or that they have not even tried.”
Roudi, who was born in Iran, has never been to a NeurIPS conference, partly because it’s difficult to obtain a visa to travel to conferences in the U.S. “As an Iranian, I have to spend a lot of time applying for a visa,” he says. “I don’t think any of these conferences are worth doing that.”
Although Roudi acknowledges that he doesn’t want conference organizers to get in trouble for not following laws, he notes that inclusivity is a “big principle of the scientific enterprise” that conferences should try to implement. If that means shifting conferences out of the U.S., Roudi says he would be fine with that. “In general, it is best to arrange events in places that are less likely to lead to exclusion,” he says.
Zador agrees. “As the U.S. becomes more authoritarian, that does seem like the natural solution to a lot of problems,” he says.
Before the 27 March announcement, some researchers had already started boycotting NeurIPS. For instance, Tan Zhi Xuan, a computer scientist at the National University of Singapore, announced on social media on 26 March that she wouldn’t referee any papers for NeurIPS this year. “While I am not affiliated with any institution on the list, or a citizen of any sanctioned country, I believe this is an absurd policy that unfairly harms researchers based on arbitrary determinations by the U.S. government,” Xuan wrote at the time.
After the announcement went live, Xuan welcomed the NeurIPS organizers clarifying their stance. But, she added on X on 27 March, “Still don’t like living in a world where academic publishing is subject to unilateral U.S. control.”
It’s not the first time sanctions have affected scientific activities. In 2004, for instance, OFAC reversed a ruling that would have required academic journals to obtain the U.S. government’s permission to edit papers authored by researchers based in countries that are under trade embargoes. At the time, the decision meant that journals run by the Institute of Electrical and Electronics Engineers could continue peer-reviewing and copyediting research papers produced in Iran, Iraq, Cuba, Libya and Sudan.
NeurIPS has also faced other challenges in recent years. The conference, for instance, saw submissions more than double from just under 9,500 in 2020 to more than 21,500 in 2025. An analysis also revealed that many of the papers accepted by the conference in 2025 contained hallucinated citations—a telltale sign that generative AI was used to fabricate references.
